Black Friday 2026: which metrics to track during and after

Black Friday metrics are divided into two groups: the ones you look at every day while the campaign is live, and the ones that only make sense after the peak has passed. Whoever confuses the two groups makes two common mistakes. One is getting stuck on the day's volume and losing sight of whether that sale will repeat. The other is waiting for the whole year to end to analyze anything, when part of the reading could have happened during the campaign week itself.

This article continues the data checklist for Black Friday: there, the base was made ready, with a naming standard, an end-to-end integration test and a dashboard built before October. This one is about what to do with that dashboard when the volume actually arrives, during the peak days and in the following weeks.

Black Friday metrics to track during the peak

During the peak days, the Black Friday metrics that matter are the ones that fit in a daily check: volume, cost and quality, always broken out by channel. Without that breakdown, a consolidated number only shows that the campaign is running, not whether it is running well.

• Volume by channel: leads or orders coming in, compared hour by hour with the same channel the day before.
• Cost per result by channel: how much each lead or sale is costing, channel by channel, to reinforce or pause investment while the campaign is still running.
• Utilization rate or average ticket: the quality metric that keeps the team from celebrating volume that does not become sales.
• Source showing up as direct traffic or without UTM: a sign that a piece fell outside the tracking standard, and there is still time to fix it.

What the four have in common is comparison by channel, never the consolidated number. A good overall result can hide an extremely expensive channel offset by a cheap one. If the last item appears, it is worth running a quick check on the source of the problem, in the spirit of the data hygiene routines, before the campaign's volume piles up more crooked records.

Black Friday metrics to track after the peak

After the campaign goes off the air, the Black Friday metrics that decide whether the result was good are repurchase, returns and adjusted CAC, because only they say whether the sale held up. The campaign day's volume tells one part of the story. The other part shows up in the following weeks.

• Repurchase rate at 30 and 60 days: how many Black Friday customers bought again, and through which channel. It is the first sign that the discount brought in customers, not just one-off orders.
• Return and cancellation rate: a canceled or returned order is not a sale, and it needs to come out of the average ticket when it is recalculated after the exchange period.
• CAC adjusted for cancellation: the same acquisition cost calculation, but with the customers who returned their purchase left out. It tends to be higher than the CAC on campaign day.
• Comparison with the same period last year: without this reference, every number from the date looks good, because volume goes up on Black Friday regardless.

These four metrics rarely appear in the campaign-day report, because they depend on a waiting period to exist. Repurchase only shows up in December or January, and returns follow the product's exchange period. That is why the Black Friday reading continues after the date has passed, within the same relationship between average ticket and revenue by channel that supports the rest of the year.

One caution before comparing the cost per result of these weeks: cutting campaign budgets as soon as Black Friday ends can make the cost per result rise on its own, without the channel having gotten worse. The Google Ads documentation explains that a significant change restarts the campaign's learning period, and the bidding algorithm needs a few conversions to adjust again.

Resource · In PortuguesePrompt Guide for Marketing and Sales70 prompts with questions about marketing, sales and behavior to ask your data.Download the guide

How to compare the result without fooling yourself

Comparing Black Friday with the previous month is the most common mistake, because any channel grows in a campaign month. The reference that supports a decision is the same period last year, channel by channel, inside a dashboard that already shows the two dates side by side. Without that breakdown ready, the comparison becomes manual work again.

What to do with the numbers after the peak has passed

The work after Black Friday is not just analyzing, it is deciding what to do with those who bought and those who did not. Two concrete actions come straight out of these metrics.

• Those who bought and did not return within 60 days enter a repurchase flow, with a different offer from the campaign. The discount that brought the first order does not need to repeat to bring the second.
• Those who clicked or got in touch and did not buy become a re-engagement lead, with the channel and campaign of origin preserved to measure whether that re-engagement actually works.

Without metrics turned into indicators, with a reference and a defined owner, each of these actions is left loose. It is the same organized base that supports both things: the dashboard that answered during the peak is the one that later segments who buys again and who needs a push.

Frequently asked questions

What is the difference between the during and after Black Friday metrics?

The during metrics answer whether the campaign is running well that day, such as volume, cost and average ticket by channel. The after metrics answer whether the sale held up, such as repurchase, returns and adjusted CAC, and they only exist after a period of days or weeks.

How long after Black Friday is it worth tracking repurchase?

Thirty and sixty days are the most useful windows. At 30 days you can already see the first sign of repurchase; at 60 days the number becomes more reliable, because it covers those who take a little longer to return.

Why can the cost per result rise after the campaign, even with nothing wrong?

Cutting or pausing budget all at once restarts the campaign's learning period on the media platform, and the bidding algorithm needs a few conversions to readjust. The fluctuation in those first days does not necessarily mean the channel got worse.

Do I need to compare with last year even if the operation grew a lot?

You do, and even more so in that case. Without the historical comparison, real growth is hidden behind the natural growth of the date, and the operation cannot separate what was the campaign's merit from what would have happened anyway.